All right, I was being flippant. But I have heard this "rates are really low zomg it's stupid not to borrow a lot and stimulate the economy!" from left-leaning commentators all over the place. I think this is a perilous idea.
1) Stimulus has a pretty mixed track record historically. Whether you buy into Ricardian equivalence or not (I suspect you do not), there are significant examples of debt-financed stimulus basically failing at the task of stimulating the economy (ie Japan). Now there are plenty of articles being written about how "oh but Japan did this and that and the other thing wrong, but we can do much better." But you know already how I feel about the optimistic view of the competence of government. Also for example see http://www.economics.harvard.edu/faculty/barro/files/Barro%2BRedlick%2Bpaper%2B0210.pdf for an empirical study of GDP responses to government spending. This is a hard topic to study, so the results of that paper are ultimately a little unsatisfying, but whatever.
2) Say the ten-year Treasury is at 2%. That's not forever. In fact, it's not even a very long time. And there is a lot of reason to believe that rates will be a lot higher in ten years, when it comes time to roll over that debt. So the bar for "we should borrow" isn't set at "we can make 3% and borrow at 2%." It should be much higher than that. That should be the central lesson of the recent credit crisis, that businesses (or governments) who are heavily leveraged in short-term debt are extremely vulnerable.
3) Accumulation of debt is cumulative. We are going to incur a lot more debt in the next few decades, because we lack the political will to do something about it. One of the great dangers of bank crises is the rapid accumulation of domestic and external debt associated with reduction in revenues. Now a Keynesian would say "we need to stimulate in times of recession and run surpluses in boom times." But again that pesky political will thing. With huge health care expenditures looming in the pretty-near future, we would do well to try not to weaken our ability to borrow in the future for any but the clearest of reasons.
4) Whatever we do, the economy isn't going to get a lot better in the short to medium-term. A few reasons for this: a) The economy was never really that good. The "good economy" of the 2000s was fueled by lax lending standards, a real estate asset bubble and a corresponding personal credit bubble that will take years to recover. b) Tax cuts will be non-stimulative in the current environment. This credit bubble seriously impacts marginal propensity to consume, because individuals are far overextended due to lax lending standards and illusory wealth from housing prices. Hence extra money in the pockets of individuals will tend to go toward debt repayment and not consumption. In this way, the US situation is a lot like Japan's, where the marginal propensity to consume is very low for quite different reasons. It is for this reason that I think that tax cuts are poor policy now. c) Lending standards have swung from overly lax to overly tight, both as the result of higher capital requirements and, well, in some cases, reasonableness. The Fed has very little room to encourage lending; they've pretty much put the jets on full and there's not a whole lot more to do. But bank lending is extremely important to the small business economy (the ones that don't have access to capital markets), and job creation, etc.
So basically my position is that the US is in the preliminary stages of a very real debt crisis that will take shape over the next decade. (not the stupid kind of debt crisis where they argue about $10 in cuts). The idea of "wow we can borrow really cheaply and invest in things we think are awesome" has certain parallels in my mind to subprime borrowers circa 2005.
2) Say the ten-year Treasury is at 2%. That's not forever. In fact, it's not even a very long time. And there is a lot of reason to believe that rates will be a lot higher in ten years, when it comes time to roll over that debt. So the bar for "we should borrow" isn't set at "we can make 3% and borrow at 2%." It should be much higher than that. That should be the central lesson of the recent credit crisis, that businesses (or governments) who are heavily leveraged in short-term debt are extremely vulnerable.
Well, I don't claim to understand exactly what the markets are saying the 10-year treasury bond rate will be ten years hence. But I imagine that it must be pretty low. I don't see any reason to assume that "there is a lot of reason" to assume the rate will be higher than what the market is saying it will be. The best guess of what it will be is what the market is predicting right now. It's obviously subject to variance, but the EV is what it is. The government is a very big entity. It should be willing to bet on that with very little concern as to variance as compared to any other entity.
Even if we assume that the borrowing has to be rolled over -- which is certainly not a given, even if you think it is a political given -- it seems to me that borrowing by the government has to be discounted like no other entity can discount it. In 10 years two things will have happened in the US -- the population will have increased, and (very likely) productivity and hence GDP/capita will have increased. Therefore that 10 year bond can be rolled over using a bigger base of more productive people as 'collateral' to pay it off.
But it doesn't have to be rolled over. It may be politically impossible, but it is certainly not unreasonable to borrow money now for ten years -- say, to build roads -- and have, e.g., an increased gas tax for years 5-10 that generates enough money to pay off the bond. So we end up borrowing money at near 0% real interest rate (for a ten year bond -- a five year bond might be negative), inject the money into the recession economy now, and pay for it later using a bigger and more productive economic base at no cost or extremely small cost for the capital itself. Seems like a win-win to me.
My understanding is that the cost of capital is an extremely important variable in determining whether projects should be undertaken. If the federal government's cost of capital is near zero, zero, or negative, and no other entity's cost of capital is even close, plus the federal government can discount based on future population and productivity growth, then it seems to me that there are likely numerous projects the federal government should be undertaking that benefit society that only it can undertake. Now, maybe there really aren't -- this seems to be part of the view you are espousing -- but I just can't see that.
The fact that some interest rates are negative as opposed to marginally positive is not a be all and end all; it just magnifies the effects and illustrates more pointedly the folly of not using the advantages the federal government has to borrow at extremely favorable terms and thereby -- in my view -- expect to have an overall +EV and +wellbeing effect on society.
Even if we are "in the preliminary states of a very real debt crisis" -- a point which I am sure many economists disagree on and which I am in no position to evaluate, it would still be necessary to balance the current suffering (economic, psychological, health) of six million people who are 'unnecessarily' unemployed (compared to a 5% rate) with the possibility of a debt crisis and its effects.
The same people who are deathly afraid of hypothetical future inflation and a hypothetical future debt crisis and the effects of these things on the economy seem to be the same people who refuse to be afraid of a hypothetical climate crisis and its effects on the economy. (Not saying you are among them)
Seems like each has to weighed in terms of the current suffering that, in the first case, is being imposed, and in the second case, the near-future term suffering that would be imposed, and the potential suffering into the future, and the probability of such things happening, and our capacity to deal with the situation technologically and politically if it does happen in the future.
This is obviously a very difficult calculation to make. I don't know what a 'debt crisis' would entail in terms of suffering. Greece has a debt crisis, and things are not good there. But Japan has a debt crisis, and while its GDP may not be growing very robustly, they seem to live in a pretty damned good society -- excellent health care, excellent life expectancy, great transportation systems, technologically advanced, etc, etc. Who really cares that they are in a "debt crisis" beyond the economists?
I would be strongly inclined, unless there was a real consensus on the horrible effects of a possible 'debt crisis' some years hence (which I don't think there is), to concentrate on relieving ongoing suffering and pushing for economic growth NOW by getting people back to work, rather than worry about possible suffering in the future.
Part of your argument seems to be that nothing we could do would significantly effect economic conditions. Clearly that is an opinion which is in dispute. My guess is whatever could be done would not be as effective as those who are loudly advocating it say it would be, and would be a lot more effective than the zero effect those who claim that nothing can be done that would have any effect claim.
Obviously if nothing can be done then nothing can be done. But if something can be done, then it is cheaper to do it with interest rates at record lows than it is otherwise. And so we should take advantage of that.
I firmly believe something can be done. Potholes have been fixed in Berkeley because of ARRA; my driving experience is a lot more pleasurable and less jarring. Perhaps my tires have been spared. The Caldicott tunnel will have a fourth hole because of ARRA, resulting a large decrease in traffic congestion. These things have value, and they create jobs. Are they the most efficient use of the money? I have no idea. But there is value that results from the investment. The money just didn't get thrown into the Pacific.
1) Stimulus has a pretty mixed track record historically. Whether you buy into Ricardian equivalence or not (I suspect you do not), there are significant examples of debt-financed stimulus basically failing at the task of stimulating the economy (ie Japan). Now there are plenty of articles being written about how "oh but Japan did this and that and the other thing wrong, but we can do much better." But you know already how I feel about the optimistic view of the competence of government.
Also for example see http://www.economics.harvard.edu/faculty/barro/files/Barro%2BRedlick%2Bpaper%2B0210.pdf for an empirical study of GDP responses to government spending. This is a hard topic to study, so the results of that paper are ultimately a little unsatisfying, but whatever.
2) Say the ten-year Treasury is at 2%. That's not forever. In fact, it's not even a very long time. And there is a lot of reason to believe that rates will be a lot higher in ten years, when it comes time to roll over that debt. So the bar for "we should borrow" isn't set at "we can make 3% and borrow at 2%." It should be much higher than that. That should be the central lesson of the recent credit crisis, that businesses (or governments) who are heavily leveraged in short-term debt are extremely vulnerable.
3) Accumulation of debt is cumulative. We are going to incur a lot more debt in the next few decades, because we lack the political will to do something about it. One of the great dangers of bank crises is the rapid accumulation of domestic and external debt associated with reduction in revenues. Now a Keynesian would say "we need to stimulate in times of recession and run surpluses in boom times." But again that pesky political will thing. With huge health care expenditures looming in the pretty-near future, we would do well to try not to weaken our ability to borrow in the future for any but the clearest of reasons.
4) Whatever we do, the economy isn't going to get a lot better in the short to medium-term. A few reasons for this:
a) The economy was never really that good. The "good economy" of the 2000s was fueled by lax lending standards, a real estate asset bubble and a corresponding personal credit bubble that will take years to recover.
b) Tax cuts will be non-stimulative in the current environment. This credit bubble seriously impacts marginal propensity to consume, because individuals are far overextended due to lax lending standards and illusory wealth from housing prices. Hence extra money in the pockets of individuals will tend to go toward debt repayment and not consumption. In this way, the US situation is a lot like Japan's, where the marginal propensity to consume is very low for quite different reasons. It is for this reason that I think that tax cuts are poor policy now.
c) Lending standards have swung from overly lax to overly tight, both as the result of higher capital requirements and, well, in some cases, reasonableness. The Fed has very little room to encourage lending; they've pretty much put the jets on full and there's not a whole lot more to do. But bank lending is extremely important to the small business economy (the ones that don't have access to capital markets), and job creation, etc.
So basically my position is that the US is in the preliminary stages of a very real debt crisis that will take shape over the next decade. (not the stupid kind of debt crisis where they argue about $10 in cuts). The idea of "wow we can borrow really cheaply and invest in things we think are awesome" has certain parallels in my mind to subprime borrowers circa 2005.
Reply
Well, I don't claim to understand exactly what the markets are saying the 10-year treasury bond rate will be ten years hence. But I imagine that it must be pretty low. I don't see any reason to assume that "there is a lot of reason" to assume the rate will be higher than what the market is saying it will be. The best guess of what it will be is what the market is predicting right now. It's obviously subject to variance, but the EV is what it is. The government is a very big entity. It should be willing to bet on that with very little concern as to variance as compared to any other entity.
Even if we assume that the borrowing has to be rolled over -- which is certainly not a given, even if you think it is a political given -- it seems to me that borrowing by the government has to be discounted like no other entity can discount it. In 10 years two things will have happened in the US -- the population will have increased, and (very likely) productivity and hence GDP/capita will have increased. Therefore that 10 year bond can be rolled over using a bigger base of more productive people as 'collateral' to pay it off.
But it doesn't have to be rolled over. It may be politically impossible, but it is certainly not unreasonable to borrow money now for ten years -- say, to build roads -- and have, e.g., an increased gas tax for years 5-10 that generates enough money to pay off the bond. So we end up borrowing money at near 0% real interest rate (for a ten year bond -- a five year bond might be negative), inject the money into the recession economy now, and pay for it later using a bigger and more productive economic base at no cost or extremely small cost for the capital itself. Seems like a win-win to me.
My understanding is that the cost of capital is an extremely important variable in determining whether projects should be undertaken. If the federal government's cost of capital is near zero, zero, or negative, and no other entity's cost of capital is even close, plus the federal government can discount based on future population and productivity growth, then it seems to me that there are likely numerous projects the federal government should be undertaking that benefit society that only it can undertake. Now, maybe there really aren't -- this seems to be part of the view you are espousing -- but I just can't see that.
The fact that some interest rates are negative as opposed to marginally positive is not a be all and end all; it just magnifies the effects and illustrates more pointedly the folly of not using the advantages the federal government has to borrow at extremely favorable terms and thereby -- in my view -- expect to have an overall +EV and +wellbeing effect on society.
Reply
The same people who are deathly afraid of hypothetical future inflation and a hypothetical future debt crisis and the effects of these things on the economy seem to be the same people who refuse to be afraid of a hypothetical climate crisis and its effects on the economy. (Not saying you are among them)
Seems like each has to weighed in terms of the current suffering that, in the first case, is being imposed, and in the second case, the near-future term suffering that would be imposed, and the potential suffering into the future, and the probability of such things happening, and our capacity to deal with the situation technologically and politically if it does happen in the future.
This is obviously a very difficult calculation to make. I don't know what a 'debt crisis' would entail in terms of suffering. Greece has a debt crisis, and things are not good there. But Japan has a debt crisis, and while its GDP may not be growing very robustly, they seem to live in a pretty damned good society -- excellent health care, excellent life expectancy, great transportation systems, technologically advanced, etc, etc. Who really cares that they are in a "debt crisis" beyond the economists?
I would be strongly inclined, unless there was a real consensus on the horrible effects of a possible 'debt crisis' some years hence (which I don't think there is), to concentrate on relieving ongoing suffering and pushing for economic growth NOW by getting people back to work, rather than worry about possible suffering in the future.
Part of your argument seems to be that nothing we could do would significantly effect economic conditions. Clearly that is an opinion which is in dispute. My guess is whatever could be done would not be as effective as those who are loudly advocating it say it would be, and would be a lot more effective than the zero effect those who claim that nothing can be done that would have any effect claim.
Obviously if nothing can be done then nothing can be done. But if something can be done, then it is cheaper to do it with interest rates at record lows than it is otherwise. And so we should take advantage of that.
I firmly believe something can be done. Potholes have been fixed in Berkeley because of ARRA; my driving experience is a lot more pleasurable and less jarring. Perhaps my tires have been spared. The Caldicott tunnel will have a fourth hole because of ARRA, resulting a large decrease in traffic congestion. These things have value, and they create jobs. Are they the most efficient use of the money? I have no idea. But there is value that results from the investment. The money just didn't get thrown into the Pacific.
Reply
Leave a comment